import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from math import radians, sin, cos, sqrt, atan2
import warnings
import folium 

warnings.filterwarnings("ignore")

from sklearn.impute import SimpleImputer
from sklearn.preprocessing import StandardScaler
from sklearn.preprocessing import OrdinalEncoder,OneHotEncoder
from sklearn.pipeline import Pipeline
from sklearn.compose import ColumnTransformer
from sklearn.impute import KNNImputer

from sklearn.model_selection import train_test_split

df = pd.read_csv("C:\\Users\\TransOrg\\Downloads\\finalTrain.csv")
df.head()

df.shape

(45584, 20)

df.describe()

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 45584 entries, 0 to 45583
Data columns (total 20 columns):
 #   Column                       Non-Null Count  Dtype  
---  ------                       --------------  -----  
 0   ID                           45584 non-null  object 
 1   Delivery_person_ID           45584 non-null  object 
 2   Delivery_person_Age          43730 non-null  float64
 3   Delivery_person_Ratings      43676 non-null  float64
 4   Restaurant_latitude          45584 non-null  float64
 5   Restaurant_longitude         45584 non-null  float64
 6   Delivery_location_latitude   45584 non-null  float64
 7   Delivery_location_longitude  45584 non-null  float64
 8   Order_Date                   45584 non-null  object 
 9   Time_Orderd                  43853 non-null  object 
 10  Time_Order_picked            45584 non-null  object 
 11  Weather_conditions           44968 non-null  object 
 12  Road_traffic_density         44983 non-null  object 
 13  Vehicle_condition            45584 non-null  int64  
 14  Type_of_order                45584 non-null  object 
 15  Type_of_vehicle              45584 non-null  object 
 16  multiple_deliveries          44591 non-null  float64
 17  Festival                     45356 non-null  object 
 18  City                         44384 non-null  object 
 19  Time_taken (min)             45584 non-null  int64  
dtypes: float64(7), int64(2), object(11)
memory usage: 7.0+ MB

df.isnull().sum()

ID                                0
Delivery_person_ID                0
Delivery_person_Age            1854
Delivery_person_Ratings        1908
Restaurant_latitude               0
Restaurant_longitude              0
Delivery_location_latitude        0
Delivery_location_longitude       0
Order_Date                        0
Time_Orderd                    1731
Time_Order_picked                 0
Weather_conditions              616
Road_traffic_density            601
Vehicle_condition                 0
Type_of_order                     0
Type_of_vehicle                   0
multiple_deliveries             993
Festival                        228
City                           1200
Time_taken (min)                  0
dtype: int64

for i in df.columns:
    print(f" no. of unique value in {i}  is   : {df[i].nunique()}")
    print("#######################################################")
    if (df[i].nunique()) < 8:
        print(f"           unique value are : {df[i].unique()}")
        print("***************************************************")
    else:
        pass

 no. of unique value in ID  is   : 45584
#######################################################
 no. of unique value in Delivery_person_ID  is   : 1320
#######################################################
 no. of unique value in Delivery_person_Age  is   : 22
#######################################################
 no. of unique value in Delivery_person_Ratings  is   : 28
#######################################################
 no. of unique value in Restaurant_latitude  is   : 657
#######################################################
 no. of unique value in Restaurant_longitude  is   : 518
#######################################################
 no. of unique value in Delivery_location_latitude  is   : 4373
#######################################################
 no. of unique value in Delivery_location_longitude  is   : 4373
#######################################################
 no. of unique value in Order_Date  is   : 44
#######################################################
 no. of unique value in Time_Orderd  is   : 176
#######################################################
 no. of unique value in Time_Order_picked  is   : 193
#######################################################
 no. of unique value in Weather_conditions  is   : 6
#######################################################
           unique value are : ['Fog' 'Stormy' 'Sandstorms' 'Windy' 'Cloudy' 'Sunny' nan]
***************************************************
 no. of unique value in Road_traffic_density  is   : 4
#######################################################
           unique value are : ['Jam' 'High' 'Medium' 'Low' nan]
***************************************************
 no. of unique value in Vehicle_condition  is   : 4
#######################################################
           unique value are : [2 1 0 3]
***************************************************
 no. of unique value in Type_of_order  is   : 4
#######################################################
           unique value are : ['Snack' 'Meal' 'Drinks' 'Buffet']
***************************************************
 no. of unique value in Type_of_vehicle  is   : 4
#######################################################
           unique value are : ['motorcycle' 'scooter' 'electric_scooter' 'bicycle']
***************************************************
 no. of unique value in multiple_deliveries  is   : 4
#######################################################
           unique value are : [ 3.  1.  0. nan  2.]
***************************************************
 no. of unique value in Festival  is   : 2
#######################################################
           unique value are : ['No' 'Yes' nan]
***************************************************
 no. of unique value in City  is   : 3
#######################################################
           unique value are : ['Metropolitian' 'Urban' 'Semi-Urban' nan]
***************************************************
 no. of unique value in Time_taken (min)  is   : 45
#######################################################

# lets create some features with the help of our existing features

feature = []
dtypes = []
count = []
unique = []
missing = []
missing_percentage=[]

for column in df.columns:
    feature.append(column)
    count.append(len(df[column]))
    missing.append(df[column].isnull().sum())
    unique.append(df[column].nunique())
    missing_percentage.append((df[column].isnull().sum() / df.shape[0]) * 100)
    dtypes.append(df[column].dtypes)
    
dataframe=pd.DataFrame({'feature':feature,
             'count':count,
             'missing':missing,
             'nunique':unique,
              'missing_percentage':missing_percentage,
              'dtypes':dtypes
              
             }) 
dataframe.set_index('feature')

df.drop('ID', axis=1, inplace=True)

df

df['Order_Date'] = pd.to_datetime(df['Order_Date'], format='%d-%m-%Y')
df.head()

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 45584 entries, 0 to 45583
Data columns (total 19 columns):
 #   Column                       Non-Null Count  Dtype         
---  ------                       --------------  -----         
 0   Delivery_person_ID           45584 non-null  object        
 1   Delivery_person_Age          43730 non-null  float64       
 2   Delivery_person_Ratings      43676 non-null  float64       
 3   Restaurant_latitude          45584 non-null  float64       
 4   Restaurant_longitude         45584 non-null  float64       
 5   Delivery_location_latitude   45584 non-null  float64       
 6   Delivery_location_longitude  45584 non-null  float64       
 7   Order_Date                   45584 non-null  datetime64[ns]
 8   Time_Orderd                  43853 non-null  object        
 9   Time_Order_picked            45584 non-null  object        
 10  Weather_conditions           44968 non-null  object        
 11  Road_traffic_density         44983 non-null  object        
 12  Vehicle_condition            45584 non-null  int64         
 13  Type_of_order                45584 non-null  object        
 14  Type_of_vehicle              45584 non-null  object        
 15  multiple_deliveries          44591 non-null  float64       
 16  Festival                     45356 non-null  object        
 17  City                         44384 non-null  object        
 18  Time_taken (min)             45584 non-null  int64         
dtypes: datetime64[ns](1), float64(7), int64(2), object(9)
memory usage: 6.6+ MB

df['year']= df['Order_Date'].dt.year
df['month']= df['Order_Date'].dt.month
df['day']= df['Order_Date'].dt.day

df.head()

df.drop('Order_Date',axis=1,inplace=True)

df.dropna(subset=['Time_Orderd'],inplace=True)

df['Time_Orderd'] = df['Time_Orderd'].str.replace('.', ':')

df['Time_Orderd'].sample(10)

10882    21:25
1257     18:15
32630    10:50
27965    17:55
11534    14:50
40590    23:30
43193    21:50
5624     10:30
15060    17:30
38883    14:30
Name: Time_Orderd, dtype: object

# Define a lambda function with error handling

def extract_time(x):
    try:
        return x.split(':')[0] + ':' + x.split(':')[1][:2]
    except IndexError:
        return '00:00'

# Apply the lambda function to the Order_Time column
df['Time_Orderd'] = df['Time_Orderd'].apply(extract_time)

# Convert the Order_Time column to a time data type
df['Time_Orderd'] = pd.to_datetime(df['Time_Orderd'], format='%H:%M:%S',errors='ignore')

df['TimeOrder_Hour']=df['Time_Orderd'].str.split(':',expand=True)[0] # 12:00 -> 12, 00


# replacing '0' with '00'  - both indicate 24th hr 
df['TimeOrder_Hour']=df['TimeOrder_Hour'].replace('0','00') #12:00

# changing the datatype
df['TimeOrder_Hour']=df['TimeOrder_Hour'].astype('int32')

# order_minute
df['TimeOrder_min']=df['Time_Orderd'].str.split(':',expand=True)[1]

# changing the datatype
df['TimeOrder_min']=df['TimeOrder_min'].astype('int32')

df.head()

df['Time_Order_picked'] = df['Time_Order_picked'].str.replace('.', ':')

# Define a lambda function with error handling
def extract_time(x):
    try:
        return x.split(':')[0] + ':' + x.split(':')[1][:2]
    except IndexError:
        return '00:00'

# Apply the lambda function to the Order_Time column
df['Time_Order_picked'] = df['Time_Order_picked'].apply(extract_time)

# Convert the Order_Time column to a time data type
df['Time_Order_picked'] = pd.to_datetime(df['Time_Order_picked'], format='%H:%M:%S',errors='ignore')

df['Time_Order_picked_Hour']=df['Time_Order_picked'].str.split(':',expand=True)[0]
# changing the datatype
df['Time_Order_picked_Hour']=df['Time_Order_picked_Hour'].astype('int32')

# order_minute
df['Time_Order_picked_min']=df['Time_Order_picked'].str.split(':',expand=True)[1]

# changing the datatype
df['Time_Order_picked_min']=df['Time_Order_picked_min'].astype('int32')

# spilliting the Delivery_person_id  on the bases of "RES", this will give us the city name of delivery

df['Delivery_city']=df['Delivery_person_ID'].str.split('RES',expand=True)[0]

df.head()

df['Delivery_city'].unique()

array(['DEH', 'KOC', 'PUNE', 'LUDH', 'KNP', 'MUM', 'MYS', 'HYD', 'KOL',
       'RANCHI', 'COIMB', 'CHEN', 'JAP', 'SUR', 'BANG', 'GOA', 'AURG',
       'AGR', 'VAD', 'ALH', 'BHP', 'INDO'], dtype=object)

df.columns

Index(['Delivery_person_ID', 'Delivery_person_Age', 'Delivery_person_Ratings',
       'Restaurant_latitude', 'Restaurant_longitude',
       'Delivery_location_latitude', 'Delivery_location_longitude',
       'Time_Orderd', 'Time_Order_picked', 'Weather_conditions',
       'Road_traffic_density', 'Vehicle_condition', 'Type_of_order',
       'Type_of_vehicle', 'multiple_deliveries', 'Festival', 'City',
       'Time_taken (min)', 'year', 'month', 'day', 'TimeOrder_Hour',
       'TimeOrder_min', 'Time_Order_picked_Hour', 'Time_Order_picked_min',
       'Delivery_city'],
      dtype='object')

df['Delivery_person_Ratings'].isnull().sum()

268

rating_map = round(df.groupby('Delivery_person_ID')['Delivery_person_Ratings'].mean(),1).to_dict()

df['Delivery_person_Ratings']=df['Delivery_person_Ratings'].fillna(df['Delivery_person_ID'].map(rating_map))
df['Delivery_person_Ratings'].isnull().sum()

0

folium.Map(tiles="cartodb positron")

import plotly.express as px

# Make a copy of the DataFrame
data1 = df.copy()

# Drop rows with any missing values
data1.dropna(axis=0, how='any', subset=['Restaurant_latitude', 'Restaurant_longitude'], inplace=True)

# Define the color scale
color_scale = [(0, 'orange'), (1, 'red')]

# Create the scatter mapbox plot
fig = px.scatter_mapbox(
    data1, 
    lat="Restaurant_latitude", 
    lon="Restaurant_longitude", 
    color_continuous_scale=color_scale, 
    zoom=8, 
    height=1000, 
    width=1000
)

# Update the map layout to use open-street-map style
fig.update_layout(mapbox_style="open-street-map")

# Remove the margins
fig.update_layout(margin={"r":0, "t":0, "l":0, "b":0})

# Show the plot
fig.show()

# latitude of india is 8°4` N  to 37°6` North
# longitude of india is 68°7` E to 97°25` East

# this data set is realted to india. but in some of latitidude is given in negative(-), negative lattitude is related
# to indian ocean ..

# so that there is much chances that negative sign in lattitude is by mistake

# changing -ve to +ve dataset
"""
abs() function return the absolute value. The absolute
value of any number is always positive it removes the negative sign of a number in Python.
"""
df['Restaurant_latitude'] = df['Restaurant_latitude'].abs()

import math

def distance(lat1, lon1, lat2, lon2):
    # Convert latitude and longitude to radians
    lat1, lon1, lat2, lon2 = map(math.radians, [lat1, lon1, lat2, lon2])
    
    # Haversine formula
    dlat = lat2 - lat1
    dlon = lon2 - lon1
    a = math.sin(dlat/2)**2 + math.cos(lat1) * math.cos(lat2) * math.sin(dlon/2)**2
    c = 2 * math.atan2(math.sqrt(a), math.sqrt(1-a))
    R = 6371.0 # Earth's radius in km
    dist = R * c
    
    return dist
df['distance'] = df.apply(lambda row: distance(row['Restaurant_latitude'], row['Restaurant_longitude'], row['Delivery_location_latitude'], row['Delivery_location_longitude']), axis=1)

df['distance'].head()

0    10.280582
1     6.242319
2    13.787860
3     2.930258
4    19.396618
Name: distance, dtype: float64

df.groupby(['Restaurant_latitude','Restaurant_longitude','Delivery_location_latitude','Delivery_location_longitude'])['distance'].max().head(40)

Restaurant_latitude  Restaurant_longitude  Delivery_location_latitude  Delivery_location_longitude
0.000000             0.000000              0.010000                    0.010000                        1.572534
                                           0.020000                    0.020000                        3.145067
                                           0.030000                    0.030000                        4.717601
                                           0.040000                    0.040000                        6.290135
                                           0.050000                    0.050000                        7.862668
                                           0.060000                    0.060000                        9.435202
                                           0.070000                    0.070000                       11.007735
                                           0.080000                    0.080000                       12.580268
                                           0.090000                    0.090000                       14.152801
                                           0.110000                    0.110000                       17.297866
                                           0.130000                    0.130000                       20.442930
9.957144             76.296783             9.967144                    76.306783                       1.560724
                                           9.977144                    76.316783                       3.121424
                                           9.987144                    76.326783                       4.682100
                                           9.997144                    76.336783                       6.242753
                                           10.007144                   76.346783                       7.803382
                                           10.017144                   76.356783                       9.363988
                                           10.027144                   76.366783                      10.924570
                                           10.037144                   76.376783                      12.485128
                                           10.047144                   76.386783                      14.045662
                                           10.067144                   76.406783                      17.166660
                                           10.087144                   76.426783                      20.287562
9.959778             76.296106             9.969778                    76.306106                       1.560717
                                           9.979778                    76.316106                       3.121411
                                           9.989778                    76.326106                       4.682082
                                           9.999778                    76.336106                       6.242728
                                           10.009778                   76.346106                       7.803351
                                           10.019778                   76.356106                       9.363950
                                           10.029778                   76.366106                      10.924526
                                           10.039778                   76.376106                      12.485078
                                           10.049778                   76.386106                      14.045606
                                           10.069778                   76.406106                      17.166591
                                           10.089778                   76.426106                      20.287481
9.960846             76.293936             9.970846                    76.303936                       1.560715
                                           9.980846                    76.313936                       3.121406
                                           9.990846                    76.323936                       4.682074
                                           10.000846                   76.333936                       6.242718
                                           10.010846                   76.343936                       7.803339
                                           10.020846                   76.353936                       9.363935
                                           10.030846                   76.363936                      10.924508
Name: distance, dtype: float64

# droping those lattitude and longitude which are less than and above to
# latitude of india is 8°4` N  to 37°6` North
# longitude of india is 68°7` E to 97°25` East

df[(df['Restaurant_latitude'] < 8) & (df['Delivery_location_latitude'] < 8)].shape

(3509, 27)

# droping  where  restaurant_lattitude and delivery_location_latitude less than 8

df = df.drop(df[(df['Restaurant_latitude'] < 8) & (df['Delivery_location_latitude'] < 8)].index)

df[(df['Restaurant_latitude'] > 37) & (df['Delivery_location_latitude'] > 37)]

df[(df['Restaurant_longitude'] < 68) & (df['Delivery_location_longitude'] < 68)]

df.columns

Index(['Delivery_person_ID', 'Delivery_person_Age', 'Delivery_person_Ratings',
       'Restaurant_latitude', 'Restaurant_longitude',
       'Delivery_location_latitude', 'Delivery_location_longitude',
       'Time_Orderd', 'Time_Order_picked', 'Weather_conditions',
       'Road_traffic_density', 'Vehicle_condition', 'Type_of_order',
       'Type_of_vehicle', 'multiple_deliveries', 'Festival', 'City',
       'Time_taken (min)', 'year', 'month', 'day', 'TimeOrder_Hour',
       'TimeOrder_min', 'Time_Order_picked_Hour', 'Time_Order_picked_min',
       'Delivery_city', 'distance'],
      dtype='object')

min_distance=df.groupby('distance')[['Restaurant_latitude','Restaurant_longitude','Delivery_location_latitude',
                        'Delivery_location_longitude']].min().sort_values(by='distance',ascending=True).head(1)
min_distance

import folium
from branca.element import Figure
fig2=Figure(width=850,height=550)

# Create a map object centered on a specific location
m = folium.Map(location=[30.914057, 75.83982], zoom_start=12)

# Add a marker for the restaurant location
folium.Marker(location=[30.914057, 75.83982], popup="Restaurant",icon=folium.Icon(color='purple',prefix='fa',icon='anchor')).add_to(m)

# Add a marker for the delivery location
folium.Marker(location=[30.924057, 75.84982], popup="Delivery").add_to(m)

# Draw a line between the two locations
folium.PolyLine(locations=[[30.914057, 75.83982], [30.924057, 75.84982]], color='red').add_to(m)

# Display the map

fig2.add_child(m)
folium.TileLayer('Stamen Terrain').add_to(m)
folium.TileLayer('Stamen Toner').add_to(m)
folium.TileLayer('Stamen Water Color').add_to(m)
folium.TileLayer('cartodbpositron').add_to(m)
folium.TileLayer('cartodbdark_matter').add_to(m)
folium.LayerControl().add_to(m)
m

max_distance=df.groupby('distance')[['Restaurant_latitude','Restaurant_longitude','Delivery_location_latitude',
                        'Delivery_location_longitude']].max().sort_values(by='distance',ascending=False).head(1)
max_distance

tooltips=f"Distance: {df['distance'].max()} km"

import folium
from branca.element import Figure
fig2=Figure(width=850,height=550)

# Create a map object centered on a specific location
m = folium.Map(location=[25.443994, 81.860187], zoom_start=12)

# Add a marker for the restaurant location
folium.Marker(location=[25.443994, 81.860187], popup="Restaurant",icon=folium.Icon(color='purple',prefix='fa',icon='anchor')).add_to(m)

# Add a marker for the delivery location
folium.Marker(location=[25.583994, 82.000187], popup="Delivery").add_to(m)

# Draw a line between the two locations
folium.PolyLine(locations=[[25.443994, 81.860187], [25.583994,82.000187]], color='red',tooltip=tooltips).add_to(m)

# Display the map

fig2.add_child(m)
folium.TileLayer('Stamen Terrain').add_to(m)
folium.TileLayer('Stamen Toner').add_to(m)
folium.TileLayer('Stamen Water Color').add_to(m)
folium.TileLayer('cartodbpositron').add_to(m)
folium.TileLayer('cartodbdark_matter').add_to(m)
folium.LayerControl().add_to(m)
m

import plotly.express as px

# Copy the DataFrame
data1 = df.copy()

# Drop rows with any missing values
data1.dropna(axis=0, how='any', subset=['Restaurant_latitude', 'Restaurant_longitude'], inplace=True)

# Define the color scale
color_scale = [(0, 'orange'), (1, 'red')]

# Create the scatter mapbox plot
fig = px.scatter_mapbox(
    data1, 
    lat="Restaurant_latitude", 
    lon="Restaurant_longitude", 
    color_continuous_scale=color_scale, 
    zoom=8, 
    height=1000, 
    width=1000
)

# Update the map layout to use open-street-map style
fig.update_layout(mapbox_style="open-street-map")

# Remove the margins
fig.update_layout(margin={"r":0, "t":0, "l":0, "b":0})

import matplotlib.pyplot as plt
import seaborn as sns

# Create a figure with specified size
plt.figure(figsize=(12, 8))

# Use sns.countplot with the DataFrame and column name
sns.countplot(x='Delivery_city', data=df)

# Rotate the x-ticks for better readability
plt.xticks(rotation='vertical')

# Show the plot
plt.show()

plt.figure(figsize=(12, 8))

# Use sns.countplot with the DataFrame and column name specified via the x parameter
sns.countplot(x='Weather_conditions', data=df)

# Rotate the x-ticks for better readability
plt.xticks(rotation='vertical')

# Show the plot
plt.show()

plt.figure(figsize=(12,8))
sns.countplot(x='Road_traffic_density', data=df)
plt.xticks(rotation='vertical')
plt.show()

plt.figure(figsize=(12,8))
sns.countplot(x='day', data=df)
plt.xticks(rotation='vertical')
plt.show()

plt.figure(figsize=(12,8))
sns.countplot(x='Vehicle_condition', data=df)
plt.xticks(rotation='vertical')
plt.show()

numerical_feature=[feature for feature in df.columns if df[feature].dtypes != 'O']
numerical_feature
df.drop('Delivery_person_ID',axis=1,inplace=True)
df.drop(['Time_Orderd','Time_Order_picked'],axis=1,inplace=True)

X = df.drop(labels=['Time_taken (min)'],axis=1)
y = df[['Time_taken (min)']]

num_col=[feature for feature in df.columns if df[feature].dtypes != 'O']
num_col

['Delivery_person_Age',
 'Delivery_person_Ratings',
 'Restaurant_latitude',
 'Restaurant_longitude',
 'Delivery_location_latitude',
 'Delivery_location_longitude',
 'Vehicle_condition',
 'multiple_deliveries',
 'Time_taken (min)',
 'year',
 'month',
 'day',
 'TimeOrder_Hour',
 'TimeOrder_min',
 'Time_Order_picked_Hour',
 'Time_Order_picked_min',
 'distance']

cat_col=[feature for feature in df.columns if df[feature].dtypes == 'O']
cat_col

['Weather_conditions',
 'Road_traffic_density',
 'Type_of_order',
 'Type_of_vehicle',
 'Festival',
 'City',
 'Delivery_city']

df[cat_col].head()

df[df['Road_traffic_density']=='Low']

Road_traffic_density=['Low','Medium','High','Jam']
Weather_conditions=['Sunny','Cloudy','Windy','Fog','Sandstorms','Stormy']

categorical_column=['Type_of_order','Type_of_vehicle','Festival','City','Delivery_city']
ordinal_encod=['Road_traffic_density','Weather_conditions']
numerical_column=['Delivery_person_Age','Delivery_person_Ratings','Vehicle_condition','multiple_deliveries',
                  'TimeOrder_Hour','distance']

# numerical pipeline

numerical_pipeline=Pipeline(steps=[
    ('impute',SimpleImputer(strategy='constant',fill_value=0)),
    ('scaler',StandardScaler(with_mean=False))
])

# categorical pipeline

categorical_pipeline=Pipeline(steps=[
    ('impute',SimpleImputer(strategy='most_frequent')),
    ('onehot',OneHotEncoder(handle_unknown='ignore')),
    ('scaler',StandardScaler(with_mean=False))
])


# ordinal pipeline

ordianl_pipeline=Pipeline(steps=[
    ('impute',SimpleImputer(strategy='most_frequent')),
    ('ordinal',OrdinalEncoder(categories=[Road_traffic_density,Weather_conditions])),
    ('scaler',StandardScaler(with_mean=False))   
])

preprocessor =ColumnTransformer([
    ('numerical_pipeline',numerical_pipeline,numerical_column),
    ('categorical_pipeline',categorical_pipeline,categorical_column),
    ('ordianl_pipeline',ordianl_pipeline,ordinal_encod)
])

X_train,X_test,y_train,y_test=train_test_split(X,y,test_size=0.20,random_state=42)

X_train=preprocessor.fit_transform(X_train)
X_test=preprocessor.transform(X_test)

from sklearn.linear_model import LinearRegression
from sklearn.linear_model import Lasso,Ridge,ElasticNet
from sklearn.svm import SVR
from sklearn.ensemble import RandomForestRegressor
from sklearn.ensemble import GradientBoostingRegressor
from sklearn.tree import DecisionTreeRegressor 
from xgboost import XGBRegressor
from sklearn.metrics import r2_score,mean_absolute_error,mean_squared_error

def evaluate_reg(true, predicted):
    r2 = r2_score(true, predicted) # Calculate r2 score
    MAE = mean_absolute_error(true, predicted) # Calculate MAE
    MSE = mean_squared_error(true, predicted) # Calculate MSE
    rmse = np.sqrt(mean_squared_error(true, predicted))
   
    return r2, MAE , MSE,rmse

from sklearn.ensemble import RandomForestRegressor, GradientBoostingRegressor
from sklearn.tree import DecisionTreeRegressor
from sklearn.linear_model import LinearRegression, Ridge
from xgboost import XGBRegressor
from sklearn.metrics import r2_score, mean_absolute_error, mean_squared_error

models = {
    "Random Forest": RandomForestRegressor(),
    "Decision Tree": DecisionTreeRegressor(),
    "Gradient Boosting": GradientBoostingRegressor(),
    "linear Regression": LinearRegression(),
    "Ridge":Ridge(),
    "XGBRegressor": XGBRegressor(), 
  
}

r2_list=[]
mse_list=[]
models_list = []

for i in range(len(list(models))):
    model = list(models.values())[i]
    model.fit(X_train, y_train) # Train model
   
    # Make predictions
    y_train_pred = model.predict(X_train)
    y_test_pred = model.predict(X_test)



    # Test set performance
    r2,MAE,MSE,rmse=evaluate_reg(y_test, y_test_pred)
    r2_train,MAE_train,MSE_train,rmse_train=evaluate_reg(y_train, y_train_pred)
    

    print(list(models.keys())[i])
    models_list.append(list(models.keys())[i])
    

    
    print('Model performance for Test set')
    print('- r2_score: {:.4f}'.format(r2))
    print('- MAE: {:.4f}'.format(MAE))
    print('- MSE: {:.4f}'.format(MSE))
    print('- rmse: {:.4f}'.format(rmse))
    r2_list.append(r2)
    mse_list.append(MSE)
    
    print('-'*35)
        
    print('Model performance for train set')
    print('- r2_score: {:.4f}'.format(r2_train))
    print('- MAE: {:.4f}'.format(MAE_train))
    print('- MSE: {:.4f}'.format(MSE_train))
    print('- rmse: {:.4f}'.format(rmse_train))

    print('='*35)
    print('='*35)
    print('\n')

Random Forest
Model performance for Test set
- r2_score: 0.8283
- MAE: 3.1358
- MSE: 15.3218
- rmse: 3.9143
-----------------------------------
Model performance for train set
- r2_score: 0.9759
- MAE: 1.1483
- MSE: 2.1068
- rmse: 1.4515
===================================
===================================


Decision Tree
Model performance for Test set
- r2_score: 0.6935
- MAE: 3.9932
- MSE: 27.3604
- rmse: 5.2307
-----------------------------------
Model performance for train set
- r2_score: 1.0000
- MAE: 0.0000
- MSE: 0.0000
- rmse: 0.0000
===================================
===================================


Gradient Boosting
Model performance for Test set
- r2_score: 0.7697
- MAE: 3.6335
- MSE: 20.5584
- rmse: 4.5341
-----------------------------------
Model performance for train set
- r2_score: 0.7740
- MAE: 3.5680
- MSE: 19.7730
- rmse: 4.4467
===================================
===================================


linear Regression
Model performance for Test set
- r2_score: 0.5351
- MAE: 5.1319
- MSE: 41.4926
- rmse: 6.4415
-----------------------------------
Model performance for train set
- r2_score: 0.5396
- MAE: 5.0371
- MSE: 40.2769
- rmse: 6.3464
===================================
===================================


Ridge
Model performance for Test set
- r2_score: 0.5352
- MAE: 5.1301
- MSE: 41.4830
- rmse: 6.4407
-----------------------------------
Model performance for train set
- r2_score: 0.5396
- MAE: 5.0352
- MSE: 40.2714
- rmse: 6.3460
===================================
===================================


XGBRegressor
Model performance for Test set
- r2_score: 0.8288
- MAE: 3.1347
- MSE: 15.2790
- rmse: 3.9088
-----------------------------------
Model performance for train set
- r2_score: 0.8752
- MAE: 2.6442
- MSE: 10.9201
- rmse: 3.3046
===================================
===================================

Results = pd.DataFrame(list(zip(models_list, r2_list)), columns=['Model Name', 'r2_score'])
Results.sort_values(by=["r2_score"],ascending=False)

Results = pd.DataFrame(list(zip(models_list, mse_list)), columns=['Model Name', 'Mse'])
Results.sort_values(by=["Mse"],ascending=True)

	Delivery_person_Age	Delivery_person_Ratings	Restaurant_latitude	Restaurant_longitude	Delivery_location_latitude	Delivery_location_longitude	Vehicle_condition	multiple_deliveries	Time_taken (min)
count	43730.000000	43676.000000	45584.000000	45584.000000	45584.000000	45584.000000	45584.000000	44591.000000	45584.000000
mean	29.566911	4.633774	17.017948	70.229684	17.465480	70.844161	1.023385	0.744635	26.293963
std	5.815064	0.334744	8.185674	22.885575	7.335562	21.120578	0.839055	0.572510	9.384298
min	15.000000	1.000000	-30.905562	-88.366217	0.010000	0.010000	0.000000	0.000000	10.000000
25%	25.000000	4.500000	12.933284	73.170000	12.988453	73.280000	0.000000	0.000000	19.000000
50%	30.000000	4.700000	18.551440	75.897963	18.633934	76.002574	1.000000	1.000000	26.000000
75%	35.000000	4.900000	22.728163	78.044095	22.785049	78.107044	2.000000	1.000000	32.000000
max	50.000000	6.000000	30.914057	88.433452	31.054057	88.563452	3.000000	3.000000	54.000000

	Delivery_person_ID	Delivery_person_Age	Delivery_person_Ratings	Restaurant_latitude	Restaurant_longitude	Delivery_location_latitude	Delivery_location_longitude	Order_Date	Time_Orderd	Time_Order_picked	Weather_conditions	Road_traffic_density	Vehicle_condition	Type_of_order	Type_of_vehicle	multiple_deliveries	Festival	City	Time_taken (min)
0	DEHRES17DEL01	36.0	4.2	30.327968	78.046106	30.397968	78.116106	12-02-2022	21:55	22:10	Fog	Jam	2	Snack	motorcycle	3.0	No	Metropolitian	46
1	KOCRES16DEL01	21.0	4.7	10.003064	76.307589	10.043064	76.347589	13-02-2022	14:55	15:05	Stormy	High	1	Meal	motorcycle	1.0	No	Metropolitian	23
2	PUNERES13DEL03	23.0	4.7	18.562450	73.916619	18.652450	74.006619	04-03-2022	17:30	17:40	Sandstorms	Medium	1	Drinks	scooter	1.0	No	Metropolitian	21
3	LUDHRES15DEL02	34.0	4.3	30.899584	75.809346	30.919584	75.829346	13-02-2022	09:20	09:30	Sandstorms	Low	0	Buffet	motorcycle	0.0	No	Metropolitian	20
4	KNPRES14DEL02	24.0	4.7	26.463504	80.372929	26.593504	80.502929	14-02-2022	19:50	20:05	Fog	Jam	1	Snack	scooter	1.0	No	Metropolitian	41
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
45579	JAPRES04DEL01	30.0	4.8	26.902328	75.794257	26.912328	75.804257	24-03-2022	11:35	11:45	Windy	High	1	Meal	motorcycle	0.0	No	Metropolitian	32
45580	AGRRES16DEL01	21.0	4.6	0.000000	0.000000	0.070000	0.070000	16-02-2022	19:55	20:10	Windy	Jam	0	Buffet	motorcycle	1.0	No	Metropolitian	36
45581	CHENRES08DEL03	30.0	4.9	13.022394	80.242439	13.052394	80.272439	11-03-2022	23:50	24:05:00	Cloudy	Low	1	Drinks	scooter	0.0	No	Metropolitian	16
45582	COIMBRES11DEL01	20.0	4.7	11.001753	76.986241	11.041753	77.026241	07-03-2022	13:35	13:40	Cloudy	High	0	Snack	motorcycle	1.0	No	Metropolitian	26
45583	RANCHIRES09DEL02	23.0	4.9	23.351058	85.325731	23.431058	85.405731	02-03-2022	17:10	17:15	Fog	Medium	2	Snack	scooter	1.0	No	Metropolitian	36

	Delivery_person_ID	Delivery_person_Age	Delivery_person_Ratings	Restaurant_latitude	Restaurant_longitude	Delivery_location_latitude	Delivery_location_longitude	Order_Date	Time_Orderd	Time_Order_picked	...	Vehicle_condition	Type_of_order	Type_of_vehicle	multiple_deliveries	Festival	City	Time_taken (min)	year	month	day
0	DEHRES17DEL01	36.0	4.2	30.327968	78.046106	30.397968	78.116106	2022-02-12	21:55	22:10	...	2	Snack	motorcycle	3.0	No	Metropolitian	46	2022	2	12
1	KOCRES16DEL01	21.0	4.7	10.003064	76.307589	10.043064	76.347589	2022-02-13	14:55	15:05	...	1	Meal	motorcycle	1.0	No	Metropolitian	23	2022	2	13
2	PUNERES13DEL03	23.0	4.7	18.562450	73.916619	18.652450	74.006619	2022-03-04	17:30	17:40	...	1	Drinks	scooter	1.0	No	Metropolitian	21	2022	3	4
3	LUDHRES15DEL02	34.0	4.3	30.899584	75.809346	30.919584	75.829346	2022-02-13	09:20	09:30	...	0	Buffet	motorcycle	0.0	No	Metropolitian	20	2022	2	13
4	KNPRES14DEL02	24.0	4.7	26.463504	80.372929	26.593504	80.502929	2022-02-14	19:50	20:05	...	1	Snack	scooter	1.0	No	Metropolitian	41	2022	2	14

	Delivery_person_Age	Delivery_person_Ratings	Restaurant_latitude	Restaurant_longitude	Delivery_location_latitude	Delivery_location_longitude	Weather_conditions	Road_traffic_density	Vehicle_condition	Type_of_order	...	Time_taken (min)	year	month	day	TimeOrder_Hour	TimeOrder_min	Time_Order_picked_Hour	Time_Order_picked_min	Delivery_city	distance
3	34.0	4.3	30.899584	75.809346	30.919584	75.829346	Sandstorms	Low	0	Buffet	...	20	2022	2	13	9	20	9	30	LUDH	2.930258
11	31.0	4.7	23.357804	85.325146	23.487804	85.455146	Sandstorms	Low	2	Meal	...	26	2022	3	10	22	30	22	45	RANCHI	19.618739
12	37.0	5.0	11.003669	76.976494	11.013669	76.986494	Sandstorms	Low	1	Snack	...	20	2022	3	11	8	15	8	30	COIMB	1.558132
17	35.0	4.3	21.160522	72.771477	21.250522	72.861477	Cloudy	Low	1	Drinks	...	38	2022	3	12	23	20	23	30	SUR	13.682045
19	23.0	4.8	18.514210	73.838429	18.624210	73.948429	Windy	Low	2	Buffet	...	12	2022	4	2	23	35	23	45	PUNE	16.853619
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
45569	30.0	4.9	11.025083	77.015393	11.045083	77.035393	Windy	Low	1	Snack	...	26	2022	3	11	10	25	10	40	COIMB	3.116127
45571	34.0	4.9	12.297954	76.665169	12.317954	76.685169	Sandstorms	Low	1	Snack	...	26	2022	4	5	9	10	9	15	MYS	3.109134
45573	30.0	4.6	18.562450	73.916619	18.582450	73.936619	Cloudy	Low	0	Meal	...	20	2022	4	1	0	37	9	15	PUNE	3.064267
45575	37.0	4.8	13.022394	80.242439	13.042394	80.262439	Sandstorms	Low	2	Drinks	...	20	2022	4	5	9	35	9	50	CHEN	3.104827
45581	30.0	4.9	13.022394	80.242439	13.052394	80.272439	Cloudy	Low	1	Drinks	...	16	2022	3	11	23	50	24	5	CHEN	4.657195

In ouir TimeOrder_Hour columns¶

¶

7AM -12PM - breakfash##¶

12PM -15 PM - lunc##h¶

15PM-17 PM- after lun##ch¶

17-19 - evening sn##aks¶

19-24 - Dinner¶

filling the null value with the mean of that person id¶

working with MAP¶

To calculate the distance between two points on a sphere (such as the Earth), we can use the Haversine formula¶

Splitting in X and Y¶

	ID	Delivery_person_ID	Delivery_person_Age	Delivery_person_Ratings	Restaurant_latitude	Restaurant_longitude	Delivery_location_latitude	Delivery_location_longitude	Order_Date	Time_Orderd	Time_Order_picked	Weather_conditions	Road_traffic_density	Vehicle_condition	Type_of_order	Type_of_vehicle	multiple_deliveries	Festival	City	Time_taken (min)
0	0xcdcd	DEHRES17DEL01	36.0	4.2	30.327968	78.046106	30.397968	78.116106	12-02-2022	21:55	22:10	Fog	Jam	2	Snack	motorcycle	3.0	No	Metropolitian	46
1	0xd987	KOCRES16DEL01	21.0	4.7	10.003064	76.307589	10.043064	76.347589	13-02-2022	14:55	15:05	Stormy	High	1	Meal	motorcycle	1.0	No	Metropolitian	23
2	0x2784	PUNERES13DEL03	23.0	4.7	18.562450	73.916619	18.652450	74.006619	04-03-2022	17:30	17:40	Sandstorms	Medium	1	Drinks	scooter	1.0	No	Metropolitian	21
3	0xc8b6	LUDHRES15DEL02	34.0	4.3	30.899584	75.809346	30.919584	75.829346	13-02-2022	09:20	09:30	Sandstorms	Low	0	Buffet	motorcycle	0.0	No	Metropolitian	20
4	0xdb64	KNPRES14DEL02	24.0	4.7	26.463504	80.372929	26.593504	80.502929	14-02-2022	19:50	20:05	Fog	Jam	1	Snack	scooter	1.0	No	Metropolitian	41

	count	missing	nunique	missing_percentage	dtypes
feature
ID	45584	0	45584	0.000000	object
Delivery_person_ID	45584	0	1320	0.000000	object
Delivery_person_Age	45584	1854	22	4.067217	float64
Delivery_person_Ratings	45584	1908	28	4.185679	float64
Restaurant_latitude	45584	0	657	0.000000	float64
Restaurant_longitude	45584	0	518	0.000000	float64
Delivery_location_latitude	45584	0	4373	0.000000	float64
Delivery_location_longitude	45584	0	4373	0.000000	float64
Order_Date	45584	0	44	0.000000	object
Time_Orderd	45584	1731	176	3.797385	object
Time_Order_picked	45584	0	193	0.000000	object
Weather_conditions	45584	616	6	1.351351	object
Road_traffic_density	45584	601	4	1.318445	object
Vehicle_condition	45584	0	4	0.000000	int64
Type_of_order	45584	0	4	0.000000	object
Type_of_vehicle	45584	0	4	0.000000	object
multiple_deliveries	45584	993	4	2.178396	float64
Festival	45584	228	2	0.500176	object
City	45584	1200	3	2.632503	object
Time_taken (min)	45584	0	45	0.000000	int64

	Model Name	r2_score
5	XGBRegressor	0.828819
0	Random Forest	0.828340
2	Gradient Boosting	0.769671
1	Decision Tree	0.693463
4	Ridge	0.535238
3	linear Regression	0.535130

	Model Name	Mse
5	XGBRegressor	15.278972
0	Random Forest	15.321797
2	Gradient Boosting	20.558401
1	Decision Tree	27.360392
4	Ridge	41.482991
3	linear Regression	41.492630

	Restaurant_latitude	Restaurant_longitude	Delivery_location_latitude	Delivery_location_longitude
distance
20.969489	25.443994	81.860187	25.583994	82.000187